Figure 2. Hyperoptimism in youths decreases throughout development.

(A) Reward belief trajectories split between the age groups over trials. Children had higher reward belief (yellow) throughout the task. The line plots indicate the reward belief on every trial averaged across participants ± 1 SEM. (B) An optimism bias (as measured by the average reward belief compared to the average reward in the task) was present in all groups. Importantly, this bias weakened during development. Children (8-9 year-olds) had a higher optimism bias compared to the adolescent groups (12-13 year-olds and 16-17 year-olds), revealing the emergence of more realistic expectations during adolescence. (C) Reward belief was predicted by the intercept, previous reward magnitude, failure to exceed the threshold on the previous trial, and participant’s own belief about previous reward magnitude. The intercept, which shows the general belief about reward magnitude over trials, was lower for late adolescents (16-17 year-olds) compared to children (8-9 year-olds) and early adolescents. Additionally, the previous reward magnitude predictor was higher for late adolescents compared to children and early adolescents, suggesting increased outcome sensitivity in late adolescence. Bar plots indicate mean ± 1 SEM; *** p <.001; ** p <.01; * p <.05; yo, year-olds; au, arbitrary units.